The watchers on the wall: Inter-Korean geopolitical risk and firm risk-taking

Hoang, Khanh; Hoang, Huy Viet; Anh, Dao Le Trang Article The watchers on the wall: Inter-Korean geopolitical risk and firm risk-taking BRQ Business Research Quarterly Provided in Cooperation with: Asociación Científica de Economía y Dirección de Empresas (ACEDE), Madrid Suggested Citation: Hoang, Khanh; Hoang, Huy Viet; Anh, Dao Le Trang (2025) : The watchers on the wall: Inter-Korean geopolitical risk and firm risk-taking, BRQ Business Research Quarterly, ISSN 2340-9444, Sage Publishing, London, Vol. 28, Iss. 1, pp. 288-303, https://doi.org/10.1177/23409444231184479 This Version is available at: https://hdl.handle.net/10419/327069 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ https://doi.org/10.1177/23409444231184479 Business Research Quarterly 2025, Vol. 28(1) 288 –303 © The Author(s) 2023 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/23409444231184479 journals.sagepub.com/home/brq Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://uk.sagepub.com/aboutus/openaccess.htm). Introduction How does corporate risk-taking react to increasing geopolitical risk and threats of war? Previous studies show that corporate risk-taking is sensitive to changes in the macroeconomic and political institutions (Boubakri et al., 2013; Tran, 2019). However, little is known about how geopolitical risk affects corporate risk-taking behavior. As uncertainty in geopolitics leads to changes in the business environment and affects corporate operations (Julio & Yook, 2012; Le & Tran, 2021), firms may have to consider undertaking fewer risky operations if the political risk is high. Using a sample of South Korean firms that operate under a high degree of geopolitical risk, this study seeks to reveal how Inter-Korean geopolitical risk influences corporate risk-taking. The unique geopolitical context of the Korean Peninsula presents an interesting setting to study the impact of geopolitical risk on corporate risk-taking. In principle, North Korea and South Korea have been at war since 1950. The two parts of Korea only preserve a socalled “peace” by an armistice in 1953, which only served the purpose until the Second Korean War 1966–1969. Since the 1990s, there have been increasing concerns from South Korea and the international community as North Korea has been developing nuclear weapons. From 2006 to date, there have been six observed nuclear tests conducted in North Korea’s territory, leading to heightened geopolitical tensions in the peninsula and the Asia-Pacific region. As the threats of war and nuclear risk become more visible, it is a compelling context to study how corporates behave under extreme uncertainty. We follow Jung et al. (2021) to measure Inter-Korean geopolitical risk using a news-based approach. Because geopolitical risk is unobservable and cannot be quantified in conventional ways, Jung et al. (2021) measure this factor by counting the frequency of news articles with keyword combinations representing Inter-Korean geopolitical tensions and military events, that is, the risk associated with geopolitics between North and South Korea. Figure 1 The watchers on the wall: Inter-Korean geopolitical risk and firm risk-taking Khanh Hoang1, Huy Viet Hoang1 and Dao Le Trang Anh2 Abstract This article investigates the relationship between Inter-Korean geopolitical risk (IKGPR) and South Korean firms’ risktaking in the context of highly political tensions in the Korean peninsula. Using a recently developed news-based index of geopolitical risk, our empirical analysis shows an inverse U-shaped relationship between IKGPR and firm risk-taking in South Korea, implying that South Korean firms’ risk-taking increases with IKGPR when IKGPR is low but becomes prudent when IKGPR is high. Such an effect of IKGPR on firm risk-taking is weaker for firms that survive wartime (1900– 1969) and is stronger for firms that were established after the Second Korean War (1966–1969), suggesting the role of corporate resilience to geopolitical risk in motivating differences in corporate behaviors. Further analyses indicate that South Korean firms tend to take less risk when the nuclear risk is higher and when they are more technology-oriented. JEL CLASSIFICATION: G30, P00, O31 Keywords Corporate governance, corporate risk-taking, geopolitical risk, inter-Korea, nuclear threats, R&D intensity 1 University of Economics Ho Chi Minh City (UEH), Ho Chi Minh City, Vietnam 2RMIT University Vietnam, Hanoi, Vietnam Corresponding author: Dao Le Trang Anh, RMIT University Vietnam, Handi Resco Building, 521 Kim Ma, Ngoc Khanh, Ba Dinh, Hanoi, 100000, Vietnam. Email: [email protected] 1184479BRQ0010.1177/23409444231184479Business Research QuarterlyHoang et al. research-article2023 Regular Paper Hoang et al. 289 shows the main developments of Inter-Korean geopolitics and the movement of the Geopolitical Risk Index proposed by Jung et al. (2021). Using a multiple fixed effect model and intensive controls for firm-level, industry-level, and macro-level confounding factors during the 2002–2019 period, we find that, in general, South Korean firms’ risk-taking increases with Inter-Korean geopolitical risk when Inter-Korean geopolitical risk is low; however, this relationship reverses to form a U-shaped curve when Inter-Korean geopolitical risk is high. The empirical finding remains qualitatively unchanged after a battery of sensitivity tests. Interestingly, analysis shows that such an effect is more pronounced in firms established after the Second Korean War (1966– 1969) while remaining weaker in their counterparts. Further analyses demonstrate the variations of the impact during North Korea’s nuclear test events and among firm groups of different technology intensities and chaebols. The findings are novel and not documented elsewhere in the literature. Our study contributes to the literature in several ways. First, we show a nonlinear relationship between InterKorean geopolitical risk and South Korean firms’ risktaking, indicating the asymmetry in corporate behavior across different levels of geopolitical risk. Our finding well supports the notion that firms rationally take risk and can distinguish between taking risk and gambling (March & Shapira, 1987). Therefore, despite firms taking risk when the risk is low, they become prudent when the level of geopolitical risk exceeds a certain threshold. From there, we indicate that the differences in the research designs (linear vs nonlinear model setting) may eventually lead to contradictory findings on the relationship between geopolitical risk and corporate risk-taking. This is an important remark that provides implications for future studies in this field. Second, we suggest the role of corporate resilience to geopolitical risk in motivating differences in corporate behaviors under uncertainty. In the context of the Korean Peninsula, firms that survive wartime may develop a certain degree of resilience to the political tension between the South and the North. In contrast, firms established after the Second Korean War (e.g., firms with less resilience to political tension) exhibit stronger negative reactions to unfavorable developments in Inter-Korean politics. This finding is intuitive as South Korean firms established and survived the Inter-Korean war periods (1950–1953 and 1966–1969) have experience dealing with extreme uncertainty during the war, thus having better corporate resilience to such extraneous shocks. As a result, negative developments in Inter-Korean geopolitical tensions do not affect their behavior as much as firms that did not experience war. Third, we provide additional evidence of how the impact varies in the cross-sections and the time dimension, thus adding new understandings of the newfound 200201 200207 200301 200307 200401 200407 200501 200507 200601 200607 200701 200707 200801 200807 200901 200907 201001 201007 201101 201107 201201 201207 201301 201307 201401 201407 201501 201507 201601 201607 201701 201707 201801 201807 201901 201907 0 50 100 150 200 250 300 350 InterKorea Second nuclear Fih nuclear test Kaesong Industrial Complex shutdow Interconnenal ballisc missile launching Threats of "Christmas gi" First nuclear test Inter-Korean summit Armisce renounciaon Figure 1. Developments in Inter-Korean geopolitics and the IKGPR index during the 2002–2019 period. 290 Business Research Quarterly 28(1) relationship. The finding suggests the role of research and development (R&D) intensity in corporate operations under extreme geopolitical risk, where technologically intensive firms take less risk than their counterparts when the threats from the North draw near. As investments in R&D activities are usually regarded as risky operations (Xu et al., 2019), firms with more R&D intensity may have less incentive to take more risk in their operations under increased geopolitical risk relative to less R&D-intensive firms. The finding provides important implications for the decision-making of R&D-intensive firms under extremerisk situations. The rest of the article proceeds as follows. Section “Literature background and hypothesis development” reviews related literature and develops hypotheses. Section “Methodology and data” presents the research methodology and data used in this study. Section “Empirical results and discussion” reports and discusses empirical results. Section “Conclusion, limitations, and future research” concludes this article. Literature background and hypothesis development Some particular areas are more complicated than others due to their historical events that motivate an intense atmosphere in the areas. One of the most unpredictable regions is the Korean Peninsula, where two countries of North Korea and South Korea co-exist. The tension in the Inter-Korea relationship, as elaborated by Jung et al. (2021, p. 7), is due to their “international sanctions against North Korea, bilateral and multilateral talks to seek reconciliation, and, finally, economic cooperation between South and North Korea.” The tightening economic and financial sanctions from the international organizations and the United States threaten North Korea’s economy and at some points trigger aggressive reactions from North Korea, which, in turn, elevate tension on the geopolitical landscape of the Korean Peninsula. Although the relationship between North and South Korea has improved recently, marked by the historical handshake of President Kim Jong Un and President Moon Jae In at the shared border of the two countries, their strain is still there to solve. Different from commonly used proxies for uncertainty such as (economic) policy uncertainty and political uncertainty, geopolitical risk is overlooked in firm-level research. The scarcity of research on Inter-Korean geopolitical risk is probably partly because of a lack of proper measurements for this term alongside the complexity of the bilateral relationship. To effectively grasp the idea of geopolitical risk, it is important to understand that risk is associated with a known outcome, while uncertainty refers to the shortage of reliable information to come up with an effective prediction of the outcome (Kobrin, 1979). March and Shapira (1987, p.1404) make a point that managers “make a sharp distinction between taking risks and gambling.” According to Pratt (1964) and other early treatments, if an individual stands between two options with similar expected values, in which one’s outcome is known for certain and the other is a gamble, the person will lean toward the former option. If firms may consider trading off risks for potential future profits when the outcome is known, they tend to be more prudent with business decisions when they cannot anticipate the consequence. Hence, firms’ reactions to risk and uncertainty may not be identical. Despite that risk and uncertainty differ in nature, the antagonistic history of North and South Korea implies that geopolitical risk may bring forth geopolitical uncertainty (i.e., an increasing probability of an armed conflict leading to unpredictable business prospects) when politically risky events incessantly take place. Therefore, rising InterKorean geopolitical risk also heightens uncertainty threats in this specific context. Despite corporate responses to Inter-Korean geopolitical swings being under-studied, several papers on stock performance in the face of Inter-Korean geopolitical fluctuations have yielded dissimilar results. Jung et al. (2021) use their self-constructed Inter-Korean geopolitical risk index, based on machine learning techniques, to reveal an inverse association between geopolitical risk and stock returns of South Korean firms. In contrast, Pyo (2021) points out that the market reacts favorably when the geopolitical situation in the Peninsula turns milder; however, the market shows limited attention to negative Inter-Korea events. Moreover, how this type of risk impacts South Korea’s macroeconomic indicators also remains inconclusive due to contradictory empirical findings (Ha et al., 2022; Pyo, 2021). The divergence in empirical findings may stem from different measurement constructions; however, another possible scenario is the empirical conflicts could be an indication of variations in the impact of InterKorean geopolitical risk at different risk levels, which has not been tapped into by existing studies. Given the looming risk of geopolitics and war in the Korean Peninsula for almost a century, the latter explanation is not irrational. Although this inherent risk may somehow alter the business practice of South Korean firms, it is well noted that most currently operating firms were established after the Second World War; therefore, they are probably born with a tolerance of Inter-Korean geopolitical swing. In other words, South Korean firms may not strongly react to minor fluctuations in geopolitical risk unless when it considerably elevates. In that light, South Korean firms likely pursue some risky projects until the risk level reaches a threshold. Theoretically, Kahneman and Tversky (1979) indicate that when an agent’s utility is based on gains or losses rather than final income levels, there is a tendency toward risk-seeking behavior in the domain of minor losses. Adding to that, French and Sichel (1993) suggest that because heightened uncertainty is Hoang et al. 291 often a result of negative shocks, the adverse effect of uncertainty tends to dominate when uncertainty is high. Conversely, there is room for a positive effect when uncertainty is low. These theoretical viewpoints imply that the association between corporate risk-taking and geopolitical risk is not necessarily linear. Whereas no prior research directly delves into the linkage between corporate risk-taking and Inter-Korean geopolitical risk, current studies provide some hints at a nonlinear impact of uncertainty from the investment and cash-holding channels. Given that Inter-Korean geopolitical risk serves as a source of uncertainty in the Korean Peninsula, it is logical to associate Inter-Korean geopolitical risk with uncertainty when studying Korean firms’ risk-taking. The term risk-taking is highly linked with investment, referring to a firm’s investment decision to trade off anticipated cash flows for uncertain risks (Acharya et al., 2011). Sarkar (2000), although agrees with the validity of the real options theory, also points out that uncertainty does not always increase the value of the option to wait. Instead, higher uncertainty may also bring up the probability of investment; however, this positive effect seems to be restricted only within a low-uncertainty domain and reverses when uncertainty exceeds a certain threshold. Therefore, according to Sarkar (2000), the relationship between investment and uncertainty should follow an inverted U curve. The nonlinear effect of uncertainty on investment is also shown in Bo and Lensin (2005), an empirical study on a sample of Dutch non-financial firms. Particularly, they find that when the uncertainty is low (high), an increase in uncertainty leads to higher (lower) investments. From the cash-holding perspective, Su et al. (2020) reveal that when uncertainty emerges, firms tend to trade off their cash holdings to seize new investment opportunities to boost their earnings. However, once uncertainty is clearly present, a risk-aversion attitude prevails as continuously heightened uncertainty makes it challenging for corporate managers to assess the company’s ability to handle the risks, leading to a more conservative investment approach and an increase in the amount of cash kept for precautionary reasons (Kotcharin & Maneenop, 2020; Lee & Wang, 2021). These findings inform that firms lean toward risky decisions when uncertainty rises as long as it has not reached a certain threshold. Firms’ ability to take risks during escalated uncertainty is also restrained by the capital market. As uncertainty heightens, banks will be more prudent in lending decisions, leading to a shortage of cash supply limiting firms’ available funds for investment (De Nicolò et al., 2010; Hu & Gong, 2018). In that context, banks’ precautionary policies when geopolitical uncertainty turns visible constrain firms’ financing options (Brogaard & Detzel, 2015; Zhang et al., 2015). The increasing risk of financing gradually decimates the expected benefit from investments, making investment options unjustifiable and encouraging managers to adopt a more conservative approach (Bernanke, 1983; Gormley & Matsa, 2016; Gulen & Ion, 2016; Shapira, 1995). This exogenous impact incentivizes firms to minimize their exposure to uncertainty in the fear of the unanticipated adverse effect caused by growing uncertainty. Corporate risk-taking, hence, decreases if uncertainty is continuously magnified as a result of geopolitical swing. Firms’ risk-taking, however, may increase in the short run because the benefit gained from investment still outweighs its cost and banks do not necessarily adopt stringent measures against unnoticeable uncertainty. The different corporate risk-taking behavior during different periods of uncertainty indicates that the relationship between firms’ risk-taking and uncertainty is likely to be nonlinear. Since Inter-Korean geopolitical risk acts as a form of uncertainty imposed on the South Korean economy, we conjecture that South Korean firms demonstrate different risk-taking attitudes across escalating geopolitical tension and calm periods. Following this conjecture, we propose the research hypothesis as follows: Hypothesis 1: The relationship between Inter-Korean geopolitical risk and firm risk-taking is inverted U-shaped. Methodology and data Variable measurements Inter-Korean geopolitical risk. We use the Inter-Korean geopolitical risk index proposed by Jung et al. (2021) as our variable of interest in this study. According to Jung et al. (2021), the geopolitical risk involves four key drivers that link to each other and create a geopolitical background for the Korean Peninsula. The drivers include the conflict between the South and North Korean military, international embargoes on North Korea, bilateral and multilateral conversations to reconcile, and cooperation between South and North Korea’s economies. Jung et al. (2021) construct the Inter-Korean geopolitical risk index based on the news articles of 18 broadcasters and newspapers of BigKinds, a news-analyzing firm formed by the Korea Press Foundation. The topics of the news focus on economics, international relationships, and politics. Then, a list of keywords on headlines or contents is determined. From the search for keywords, Jung et al. (2021) calculate the news frequency in each category (military issues, sanctions, conversation of reconciliation, and economic cooperation) and divide it into two groups: positive and negative news. Positive and negative news indicate the news that reports the tension decrease and increase between South and North Korea, respectively. Let Njt be the total number of news from media j at time t. The relative frequency () Xjt of negative news () , Nnegat jt and 292 Business Research Quarterly 28(1) positive news () , N posit jt over the total numbers of news Njt is computed as follows X NN N jt negat jt posit jt jt   ,, Jung et al. (2021) then transform Xjt to a positive number as follows  XX jt jt   05 01 2 .. For each media source, Jung et al. (2021) standardize  Xjt by stdj (the time series within newspaper standard deviation of  Xjt) to get a series Yjt YX std jt jt j =  Yjt is averaged across K media sources as follows YKY t j N jt    1 1 Finally, Yt is normalized to get the Inter-Korean geopolitical risk index IKGPRt with the mean value of Y () Y IKGPRt t Y Y =100 The higher the value of IKGPR, the higher the geopolitical risk between North and South Korea is. Corporate risk-taking. Following the commonly accepted practice in the previous literature (Kim et al., 1993; Wright et al., 2007; Yung & Chen, 2018), we use the rolling standard deviation of return-on-assets ratio (SDROA) as a measurement of corporate risk-taking. SDROA is calculated over a rolling window of three continuous years. For robustness check, we compute an alternative measure of risk-taking: the three-year rolling standard deviation of return-on-common equity ratio (SDROE). The higher the value of SDROA and SDROE, the riskier projects the firm takes on and the more volatile its income is. Using an alternative approach to measure firm risktaking (Martins, 2020), we proxy firm risk-taking by the absolute deviation from expected performance for each year. Martins (2020) models the expected firm performance using the following equation ROASIZELEVERAGE CASH INVESTMENT GR it it it it it ,, , ,,         OOWTH FIXED ASSETS it it iit ,, , _    (1) where ROAit, is the return-on-total assets ratio, SIZEit, is the natural logarithm of the book value of total assets, LEVERAGEit, is the total debt to total assets ratio, CASHit, stands for cash and cash equivalents scaled by net total assets, INVESTMENTit, is the ratio of the capital expenditure on the beginning balance of total assets, GROWTHit, is the annual changes in sales scaled by 1-year lagged sales, FIXED ASSETSit _, is the tangible assets scaled by total assets ratio, δ i is the firm-fixed effect, all numbers are from firm i during year t. The absolute value of the residual ε it, is defined as the deviation of actual ROA from the expected ROA, thus serving as the alternative measure of firm risk-taking (ABS_ROA_RES). The higher the value of ABS_ROA_RES, the more risktaking the firm is. Similar approaches are adopted in the literature (Adams et al., 2005; Cheng, 2008; Nakano & Nguyen, 2012). Control variables. Based on the previous literature (Tran, 2019; Yung & Chen, 2018), we adopt country-level, industry-level, firm-level, and corporate governance controls to capture the specific characteristics of each firm in each year of our sample. Regarding country-level controls, we employ South Korea’s Economic Policy Uncertainty Index (Baker et al., 2016) and the gross domestic product (GDP) growth rate (in percentage) of South Korea. The country-level controls indicate the military issue, economic, and political background of South Korea. In terms of industry-level controls, the standard deviation of total sales of firms in the same industry and the standard deviation of ROA of firms in the same industry are included. For the firm-level controls, we select popularused control variables in the literature, including firm size, firm age, leverage, sales growth, cash ratio, fixed assets ratio, cash flow, Z-score, market-to-book ratio, and annual stock returns following Yung and Chen (2018), Tran (2019), and Koirala et al. (2020). Empirical model We use the following empirical model to investigate the impact of Inter-Korean geopolitical risk on firm risk-taking RISK TAKING IKGPRSQ IKGPRCONTROL ij t ti t itit __ , , ,       (2) Hoang et al. 293 where RISK TAKINGij _ , is the measure of firm i’s risktaking during period t; IKGPRt and β IKGPRSQt _ are the Inter-Korean Geopolitical Risk Index and its square during period t, respectively; Σ CONTRO Lit, is the vector of control variables at firm-level, industry-level, and macro-level during period t; δ i and θ t are the industryand year-fixed effects, respectively; ε it, is the error term of the model. When estimating the model, we cluster standard errors by firm to alleviate the potential impact of heteroskedasticity and autocorrelation. In general, the use of a variable and its squared term in the same regression model does not affect their p values (Alisson, 2012), and the inferences can be safely interpreted as long as the Table 1. Variable descriptions. Variable Description Sources ROA Net income scaled by average total assets (in %) Bloomberg ROE Net income scaled by average common equity Bloomberg SDROA 3-year rolling standard deviation of ROA following Yung and Chen (2018) Bloomberg SDROE 3-year rolling standard deviation of ROE following Yung and Chen (2018) Bloomberg ABS_ROA_RES The on-year risk-taking measure which equals the absolute value of the deviation from the expected performance of firm following Martins (2020) IKGPR The annualized Inter-Korean Geopolitical Risk Index calculated by taking the mean of monthly Inter-Korean Geopolitical Risk Index during a year, then scaled by 100 Jung etal. (2021) IKGPR_SQ The square of IKGPR Jung etal. (2021) MA3_IKGPR 3-year moving average of IKGPR Jung etal. (2021) MA3_IKGPR_SQ The square of MA3_IKGPR IW_IKGPR Increasing-weighted IKGPR calculated by assigning increasing weighs of 1–12 for months of the year from January to December, respectively, then taking the weighted average and scaled by 100 Jung etal. (2021) IW_IKGPR_SQ The square of IW_IKGPR DW_IKGPR Decreasing-weighted IKGPR calculated by assigning decreasing weighs of 12–1 for months of the year from January to December, respectively, then taking the weighted average and scaled by 100 Jung etal. (2021) DW_IKGPR_SQ The square of DW_IKGPR Res_IKGPR The residuals from the regression of IKGPR on the Geopolitical risk (GPR) index of China and Russia (Caldara and Iacoviello, 2022). China’s and Russia’s GPR indexes are scaled by 100 Jung etal. (2021), Caldara and Iacoviello (2022) Res_IKGPR_SQ The square of Res_IKGPR SIZE Natural logarithm of total assets Bloomberg LEVERAGE Total debt scaled by total assets Bloomberg GROWTH Growth in sales scaled by 1-year lagged sales Bloomberg CASH Cash and cash equivalents scaled by total assets Bloomberg FIXED_ASSETS Property, plant, and equipment scaled by total assets Bloomberg CASHFLOW Net operating cash flow scaled by total assets Bloomberg ZSCORE Altman’s Z-score of the firm Bloomberg MTB Market-to-book ratio Bloomberg SRETURN Annual stock returns of the firm Bloomberg FIRM_AGE Age of the firm (from incorporation to date) Bloomberg SD_SALES_IND Standard deviation of total sales of firms in the same industry Bloomberg SD_ROA_IND Standard deviation of ROA of firms in the same industry Bloomberg KEPU South Korean’s Economic Policy Uncertainty Index Baker etal. (2016) GDP GDP growth of South Korea (in %) World Bank NUCLEAR The number of North Korea’s nuclear tests per year GDP: gross domestic product. estimates are statistically significant or if the upper and lower limits of the confidence interval are in their theoretical range (O’brien, 2007). Following the hypotheses development, we expect the coefficient β of IKGPR_SQ to be negative and significant. Variable descriptions are presented in Table 1. Data and sample We collect data from several sources. Financial data of South Korean firms are from the Bloomberg database. We retrieve macroeconomic data from World Bank open database; the economic policy uncertainty index of South 294 Business Research Quarterly 28(1) Korea (Baker et al., 2016) is from the website www.policyuncertainty.com. For independent variable, we adopt the Inter-Korean Geopolitical Risk Index proposed by Jung et al. (2021). Our initial sample includes all firms listed on Korea Stock Exchange from 2002 to 2019. We screen the sample and exclude all financial and utility firms because the risk-taking nature is different from that of non-financial firms. After excluding missing data, the final sample consists of 1,668 South Korean firms with a total of 15,412 firm-year observations. All continuous variables are winsorized at the 1st and the 99th percentiles to alleviate the potential impacts of outliers on the outcomes of data analysis. Empirical results and discussion Descriptive statistics The variables’ descriptive statistics are displayed in Table 2. The average corporate risk-taking level (SDROA) of the listed Korean firms is 4.429, with a standard deviation of 6.109. The variable IKGPR during the sample period ranges from 0.527 to 2.389, with a mean of 1.257. Figure 1 shows that the Inter-Korean geopolitical risk index reaches new peaks at the announcement of important negative news, such as the first and second nuclear tests, the Yeonpyeong Island attack. The index drops to low level when positive news, for example, Inter-Korean summit and Inter-Korean and North Korea-United States summits, are informed. Regarding the control variables, an average Koreanlisted firm has leverage and sales growth of 23.2% and 9%, respectively. The average fixed assets ratio and cash ratio are 32.8% and 10.5%, correspondingly. There are both young firms and mature firms in the sample, with an average firm age (FIRM_AGE) of 30.3 years, where the youngest firms are only 1 year old and the oldest firms are 121 years old. In terms of market performance, Korean-listed firms’ market-to-book ratio and annual stock returns are averaged at 1.5% and 15.8%, respectively. For the corporate governance factor, the statistics show that a majority (81.5%) of the Chief Executive Officer is also the chair of Korean firms’ board of directors. Besides, as shown in Table 2, during the study period (2002–2019), the average GDP growth of Korea is 3.428%. The average number of North Korea’s nuclear tests is 0.419 times per year. Table 8 in the Appendix shows the pairwise correlation matrix of our regression model’s variables. Since all correlation values are smaller than 0.5, there are no notable correlations among independent variables in the regression model. Table 2. Summary statistics. Variable Obs M SD Minimum Maximum SDROA 15,412 4.429 6.109 0.000 40.366 SDROE 15,371 10.628 16.912 0.000 110.458 ABS_ROA_RES 15,412 4.206 5.253 0.000 53.863 IKGPR_SQ 15,412 1.854 1.598 0.277 5.708 IKGPR 15,412 1.257 0.523 0.527 2.389 MA3_IKGPR 15,412 121.914 32.927 73.250 177.188 IW_IKGPR 15,412 124.158 52.281 50.974 234.000 DW_IKGPR 15,412 127.284 53.300 54.359 243.833 Res_IKGPR 15,412 11.539 39.539 −46.222 94.430 SIZE 15,412 5.480 1.566 1.756 10.119 ROA 15,412 1.156 9.925 −52.101 27.224 LEVERAGE 15,412 0.232 0.175 0.000 0.696 GROWTH 15,412 0.090 0.355 −0.645 2.514 CASH 15,412 0.105 0.128 0.001 0.945 FIXED_ASSETS 15,412 0.328 0.185 0.003 0.793 CASHFLOW 15,412 −0.004 0.090 −0.384 0.242 ZSCORE 15,412 3.511 6.176 −12.981 457.813 MTB 15,412 1.500 1.656 0.120 11.439 SRETURN 15,412 0.158 0.635 −0.776 3.295 FIRM_AGE 15,412 30.308 16.537 1.000 121.000 SD_SALES_IND 15,412 3,143.717 5,861.372 6.653 52,813.613 SD_ROA_IND 15,412 9.230 4.259 0.035 35.688 KEPU 15,412 144.888 43.217 68.64 257.362 GDP 15,412 3.428 1.412 0.793 7.725 NUCLEAR 15,412 0.419 0.644 0.000 2.000 Hoang et al. 295 Baseline results and discussion The baseline regression results of Model 2 are presented in Table 3. As shown in Table 3, the coefficient of IKGPR is positively and significantly associated with corporate risk-taking at 1% level in all three regression analyses (reduced-form, full model, and stand-alone effect of Table 3. Baseline regression. Variables (1) (2) (3) Reduced-form model Full model Stand-alone effect of IKGPR SDROA SDROA SDROA IKGPR_SQ −10.958*** (2.615) −1.341*** (0.013) IKGPR 16.299*** (4.416) 4.384*** (0.054) 0.230*** (0.057) SIZE −0.507*** (0.065) −0.507*** (0.066) ROA −0.166*** (0.017) −0.166*** (0.017) LEVERAGE 0.820 (0.546) 0.820 (0.543) GROWTH 0.584** (0.204) 0.584** (0.205) CASH 2.058** (0.786) 2.058** (0.789) FIXED_ASSETS −3.001*** (0.615) −3.001*** (0.619) CASHFLOWS −3.651** (1.525) −3.651** (1.519) ZSCORE −0.032** (0.012) −0.032** (0.012) MTB 0.611*** (0.085) 0.611*** (0.084) SRETURN −0.445*** (0.131) −0.445*** (0.132) FIRM_AGE −0.008 −0.008 (0.005) (0.006) SD_SALES_IND 0.000 (0.000) 0.000 (0.000) SD_ROA_IND 0.063*** (0.016) 0.063*** (0.016) KEPU −0.019*** (0.001) −0.010*** (0.001) GDP 0.048 (0.043) 0.198*** (0.043) Constant 3.953*** (0.126) 7.007*** (0.603) 7.403*** (0.588) Industry-fixed effect Yes Yes Yes Year-fixed effect Yes Yes Yes Observations 17,350 15,412 15,412 Adjusted R2.084 .258 .258 This table reports the regression results of our baseline regression. The dependent variable is SDROA. Column 1 reports the regression result of the reduced-form model, Column 2 reports the regression results of the full model, while Column 3 shows the estimation of the standalone effect of IKGPR on SDROA after controlling other variables. All variable definitions are presented in Table 1. Standard errors are clustered by firm. Industryand year-fixed effects are included. Numbers in parentheses are standard errors. ***, **, and * denote significance levels of 1%, 5%, and 10%, respectively. IKGPR). This result suggests that corporate risk-taking of Korean-listed firms generally increases with Inter-Korean geopolitical risk. However, the risk-taking attitude could be reverted when the geopolitical risk elevates to a certain level as documented by Bo and Lensin (2005). Furthermore, the results show that the coefficient of IKGPR_SQ is negatively and significantly related to 302 Business Research Quarterly 28(1) Martins, H. C. (2020). The Brazilian bankruptcy law reform, corporate ownership concentration, and risk-taking. Managerial and Decision Economics, 41, 562–573. Nakano, M., & Nguyen, P. (2012). Board size and corporate risk-taking: Further evidence from Japan. Corporate Governance: An International Review, 20(4), 369–387. O’brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41, 673– 690. Pandit, S., Wasley, C. E., & Zach, T. (2011). The effect of research and development (R&D) inputs and outputs on the relation between the uncertainty of future operating performance and R&D expenditures. Journal of Accounting, Auditing & Finance, 26(1), 121–144. https://doi. org/10.1177/0148558X11400583 Pratt, J. W. (1964). Risk aversion in the small and in the large. Econometrica, 32, 122–136. https://doi.org/10.1016/B9780-12-214850-7.50010-3 Pyo, D. J. (2021). Does geopolitical risk matter? Evidence from South Korea. Defence and Peace Economics, 32(1), 87–106. https://doi.org/10.1080/10242694.2020.1829937 Sarkar, S. (2000). On the investment–uncertainty relationship in a real options model. Journal of Economic Dynamics and Control, 24(2), 219–225. https://doi.org/10.1016/S01651889(99)00005-6 Shapira, Z. (1995). Risk taking: A managerial perspective. Russell Sage Foundation. Su, X., Zhou, S., Xue, R., & Tian, J. (2020). Does economic policy uncertainty raise corporate precautionary cash holdings? Evidence from China. Accounting & Finance, 60(5), 4567–4592. https://doi.org/10.1111/acfi.12674 Tran, T. Q. (2019). Economic policy uncertainty and corporate risk-taking: International evidence. Journal of Multinational Financial Management, 52, 100605. https:// doi.org/10.1016/j.mulfin.2019.100605 Tübbicke, S. (2022). Entropy balancing for continuous treatment. Journal of Econometric Methods, 11(1), 71–89. Vo, L. V., & Le, H. T. T. (2017). Strategic growth option, uncertainty, and R&D investment. International Review of Financial Analysis, 51, 16–24. https://doi.org/10.1016/j. irfa.2017.03.002 Wright, P., Kroll, M., Krug, J. A., & Pettus, M. (2007). Influences of top management team incentives on firm risk taking. Strategic Management Journal, 28(1), 81–89. https://doi. org/10.1002/smj.548 Xu, D., Zhou, K. Z., & Du, F. (2019). Deviant versus aspirational risk taking: The effects of performance feedback on bribery expenditure and R&D intensity. Academy of Management Journal, 62(4), 1226–1251. https://doi.org/10.5465/amj.2016.0749 Yeoh, S. B., & Hooy, C. W. (2022). Generation effects and managerial risk taking. Journal of Business Research, 139, 918–934. https://doi.org/10.1016/j.jbusres.2021.09.063 Yung, K., & Chen, C. (2018). Managerial ability and firm risk-taking behaviour. Review of Quantitative Finance and Accounting, 51, 1005–1032. https://doi.org/10.1007/ s11156-017-0695-0 Zhang, G., Han, J., Pan, Z., & Huang, H. (2015). Economic policy uncertainty and capital structure choice: Evidence from China. Economic Systems, 39(3), 439–457. https://doi. org/10.1016/j.ecosys.2015.06.003 Zhao, Q., & Percival, D. (2017). Entropy balancing is doubly robust. Journal of Causal Inference, 5(1), 20160010. Hoang et al. 303 Appendix 1 Table 8. Correlation matrix. Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (1) SDROA 1.000 (2) SDROE 0.758 1.000 (3) IKGPR_SQ 0.041 0.022 1.000 (4) IKGPR 0.037 0.020 0.983 1.000 (5) MA3_IKGPR 0.037 0.002 0.622 0.585 1.000 (6) IW_IKGPR 0.034 0.017 0.974 0.990 0.591 1.000 (7) DW_IKGPR 0.039 0.023 0.972 0.990 0.567 0.960 1.000 (8) RES_IKGPR 0.026 0.013 0.938 0.949 0.588 0.966 0.913 1.000 (9) SIZE −0.265 −0.152 −0.035 −0.036 −0.050 −0.033 −0.038 −0.023 1.000 (10) ROA −0.381 −0.377 −0.017 −0.015 −0.052 −0.012 −0.017 −0.011 0.223 1.000 (11) LEVERAGE 0.071 0.246 −0.059 −0.058 −0.114 −0.060 −0.054 −0.061 0.166 −0.312 1.000 (12) GROWTH 0.017 0.004 −0.020 −0.031 −0.058 −0.024 −0.037 −0.027 0.020 0.124 −0.011 1.000 (13) CASH 0.115 0.011 0.092 0.093 0.171 0.095 0.088 0.090 −0.189 0.023 −0.305 0.015 1.000 (14) FIXED_ASSETS −0.151 −0.049 −0.057 −0.059 −0.109 −0.060 −0.058 −0.065 0.195 0.016 0.381 −0.047 −0.340 1.000 (15) CASHFLOW −0.206 −0.195 0.011 0.019 −0.015 0.016 0.023 0.004 0.127 0.436 −0.295 0.010 0.132 −0.102 1.000 (16) ZSCORE −0.003 −0.079 0.066 0.065 0.099 0.064 0.065 0.060 −0.124 0.146 −0.323 0.036 0.172 −0.189 0.104 1.000 (17) MTB 0.258 0.280 0.087 0.088 0.129 0.086 0.087 0.081 −0.185 −0.186 0.002 0.078 0.178 −0.174 −0.123 0.315 1.000 (18) SRETURN −0.049 −0.034 −0.021 −0.012 −0.133 −0.025 0.002 −0.024 −0.014 0.153 −0.035 0.129 0.021 0.006 0.089 0.112 0.214 1.000 (19) FIRM_AGE −0.161 −0.097 −0.013 −0.017 −0.022 −0.015 −0.019 −0.012 0.319 0.051 0.061 −0.055 −0.182 0.177 0.030 −0.110 −0.212 −0.003 1.000 This table reports the pairwise correlation matrix of Inter-Korean geopolitical risk measures and firm-level variables used in the baseline model. Bold numbers represent the correlation coefficients that are significant at 10% level or stronger.